Experimental study of Cynara Cardunculus L. gasification in bubbling fluidised bed: agglomeration, gas, tars and fly ash
- Autores: Daniel Serrano García
- Directores de la Tesis: Sergio Sánchez Delgado (dir. tes.)
- Lectura: En la Universidad Carlos III de Madrid ( España ) en 2016
- Idioma: español
- Tribunal Calificador de la Tesis: Alberto Gómez-Barea (presid.), José María Sánchez Hervás (secret.), Filomena Pinto (voc.)
- Programa de doctorado: Programa de Doctorado en Ingeniería Mecánica y de Organización Industrial por la Universidad Carlos III de Madrid
- Materias:
- Enlaces
- Tesis en acceso abierto en: e-Archivo
- Resumen
The use of biomass as an energy resource can reduce the existing dependence on fossil fuels consumption, shifting towards a more aware environmental development. It can be also an opportunity to deal with huge amounts of solid residues such as municipal solid waste, sewage sludge or agricultural residues, obtaining valuable products from them and reducing their landfill disposal. One of the routes employed for this purpose is the thermochemical conversion and, in particular, gasification. In gasification process, the biomass is transformed into a mixture of products: non condensible gases, condensible gases, solid char and ashes by means of partial oxidation at high temperature. The non condensible fraction of gas is the main product that can be use in different applications: fuel in boilers and gas engines, or raw gas to produce hydrogen, methane or biofuels through the Fischer-Tropsch process.
There are multiple gasification technologies to transform biomass by the thermochemical route: fixed beds, moving beds, fluidized beds, etc. The good mixing and high mass and heat transfer rates make fluidized beds a good option for biomass gasification. However, there are three main operational problems that need to be considered. The first problem is the bed agglomeration which is motivated by the high alkali content in biomass and the reaction temperatures reached in the reactor. These elements react with the silicon compounds from the bed material to form low melting point silicates that act as a "glue" between particles or coat them, leading to the agglomerates and to the possible defluidization of the reactor. The second problem is the tar generation, a viscous and sticky fraction that condense on cold surfaces and may clog and block the pipes and downstream devices. The third aspect is the ash generation, which constitutes a residue that need to be treated or reused in different applications before its final disposal.
This PhD thesis studies the above three mentioned related problems of biomass gasification in a bubbling fluidized bed reactor. Cynara cardunculus L., an energy crop typical from Mediterranean regions and with a high alkali content, is used as biomass feedstock in order to test its potential for gasification in a fluidized bed. This energy crop has some advantages from other plants such as the low water irrigation or the use of lands not suitable for food purposes. Another main aspect to take into account in fluidised bed gasification, is the bed material selection. In this PhD thesis, sepiolite, a clay mineral that is commonly used as adsorbent, is proposed as bed material, checking its suitability for agglomeration, gas and tar composition, and mechanical resistance. This investigation has been performed in three experimental facilities: a lab- and a pilot-scale gasifiers, and a cold fluidized bed. Different techniques have been used to analyse the data and to characterize the products from the gasification process.
The agglomeration process has been studied by means of the analysis of the pressure fluctuation signals acquired inside a lab-scale fluidized bed. Depending on the relation between the biomass particles density and the bed material density, two clearly different behaviours are observed: jetsam and flotsam. The biomass sinks inside the bed in the first case while, in the second one, the biomass floats on the bed surface due to the higher density of the bed material. The wide band energy, the attractor comparison tool, and the standard deviation methods are used in order to detect agglomeration and, as a consequence, the defluidization of the bed. The wide band energy analysis shows that, for jetsam fuel particles, the endogenous bubbles produced by the fuel devolatilization inside the bed change the energy distribution, while for flotsam fuel particles, the cap-clinker agglomerate formed is detected by high frequencies in the power spectrum. Similar defluidization times are obtained for all tested methods, being the defluidization time of sepiolite experiments considerably higher than in the silica sand tests.
The performance of sepiolite as bed material towards gas composition and tar mitigation has been investigated in a lab-scale fluidized bed gasifier, comparing the results with the same experiments operated with silica sand. The gas produced with sepiolite as bed material has a slightly lower quality than the gas generated with silica sand. However, the tar generation is rather reduced in the sepiolite bed and the tar composition is also different among the bed materials. Sepiolite properties such as surface area and morphology have been analysed by means of specific surface area (BET) and scanning electron microscopy (SEM-EDS) before and after the experiments. The fuel behaviour and the properties of sepiolite induce the adsorption of tars and molten ashes on the sepiolite surface, leading to a much better performance in terms of tars and agglomeration. In addition, a long attrition test of 125 hours has been conducted on the sepiolite, obtaining a smaller attrition rate than other common bed materials such as alumina or dolomite.
A pilot-scale gasifier has been employed to test Cynara cardunculus L. with magnesite and olivine as bed materials in terms of gas composition and tar generation. A relatively high hydrogen content in the product gas is obtained in both cases. A positive effect of the gasification temperature is observed in the gasification parameters and efficiency. Small differences in total tar are observed between magnesite and olivine, although tar composition is very different. The benzene, toluene, ethylbenzene and xylenes fraction (BTEX) is higher for olivine while similar polycyclic aromatic hydrocarbon fraction is obtained in both bed materials. Magnesium from magnesite and olivine shows a catalytic behaviour towards tar cracking. Better gasification performance is observed with magnesite at 700 ºC and with olivine at 800 ºC.
Finally, the fly ashes from the pilot-scale gasification experiments have been analysed in terms of elemental and metal composition, sulphur and chlorine contents, and leaching behaviour. Most of the elutriated fines are retained, by far, in the first cyclone. The bed material and the reactor materials also influence the final ash composition of the fines. The reuse of these fines is quite difficult in the cement industry or as fertilizer as a consequence of the high carbon, alkali, chlorine and heavy metals contents, being the use as alternative/secondary fuel a good option due to the high energy content in the fines.
